Abstract:

Purpose – The purpose of this paper is to discuss a multiscale pricing model augmented by high order moments by combining wavelet analysis, Fama-French three-factor model and high order moments. The objective is to examine the contribution of the co-skewness and co-kurtosis systematic risks to the relationship between stock returns and Fama-French risk factors at different time-scales.

Design/methodology/approach – Exploiting the multiresolution analysis which decomposes the data set into components associated with different time-scales, the three Fama-French model augmented by the high order moments will be tested over different investments periods.

Findings – The obtained results show that the higher order moments of a distribution ameliorate the explanatory power of the Fama-French three-factor model which becomes stronger as the wavelet scale increases. Besides, the relationship between the portfolio returns and the market risk factors as well as the size and value factors depends significantly upon the considered time-horizon.

Practical implications – This paper underline the importance of the non linear market risk over different time-scales. It gives investors the occasion to update the investment periods and the portfolio management strategy according to the multiscale nature of risk and return relationships.

Originality/value – The proposed methodology uses the multiresolution analysis that shows the multiscale property of the dynamic risk and return relationship. This work gives a new insight to asset pricing researcher, fund managers and financial market investors with regard to portfolio selection and investment period.